With all the talk of high profit margins, it is worth remembering that they are nearly useless as a tool for individual stock selection. If you are an index investor, then margins (and maybe more accurately, ROEs) matter a great deal because when margins/ROEs revert to longer term norms, the market will likely suffer. But if you are instead building your own portfolio (whether it be through a quantitative screen or individual stock selection), then net profit margins haven’t mattered nearly as much as other measures like valuations across history.
In the below chart, you can see the historical excess return, by decile, for two factors: net margin (relative to companies in the same industry group) and valuation (relative to the entire market). On the left hand side (decile 1) are companies with the cheapest valuations or highest margins, and on the right side (decile 10) are those with expensive valuations or very low margins.
1963-2014
You can see right away that where valuations have been highly predictive of strong future excess return, net margins haven’t been over the past 50 years. A high margin does not a great investment make-but a cheap price often does.
Calculation Notes: deciles are run using a rolling annual rebalance. The universe is all stocks with an inflation adjusted market cap > $200MM between 1963-2014. To control for differences in margins across industries, I’ve calculated margin relative to companies in the same industry group. I’ve also excluded companies with negative earnings. Value is calculated using a variety of measures like price-to-sales, price-to-earnings, EBITDA/EV. The benchmark (against which I calculate the excess returns) is an equal weighted basket of all stocks with an inflation adjusted market cap > $200MM.
/rating_on.png "1 Star")
Loading...
Patrick;
Thanks for the interesting post. Would another way of describing your conclusion be to say that historical valuation rather than cross-sectional comparison is more strongly predictive of future returns?
Thanks!
Well, where a stock falls (that is, into which decile) is a cross-sectional determination made in each period. everything is based on where stocks sit RELATIVE to the rest of the market at any given time. Basically what that means is that you want to own stocks that look cheapest through time, and that a companies margin versus its industry hasn’t been predictive of future returns.
What does the chart look like if you don’t industry neutralise it?
Patrick - great chart. What do results look like when you include companies with negative EPS and positive EBITDA?
pretty similar, although lower margin companies do a little worse
looks very similar, I will run it if I get some time in a few weeks.